Cancer is the second leading cause of death in the United States. The complexity of most solid tumors limits the understanding of tumor biology and the development of new and improved treatments. The lack of robust in vitro and in vivo models that enable culture of primary human cancers and the reconstruction of the tumor microenvironment has hampered progress in understanding response to targeted therapeutics in real time. Most current studies rely on cancer cell line culture on plastic in 2-dimensions, or the cost and labor intensive generation of patient-derived xenograft (PDX) models in immunocompromised mice. Tumor models that more closely reflect the conditions in patients are required.